Automated Computerized Detection of Unruptured Intracranial Aneurysms in MRA images(Joint Session 1)

概要

論文の詳細を見る
The purpose of this study was to develop an automated computerized detection of aneurysms on magnetic resonance angiography (MRA) images in order to assist radiologists' interpretation as a "second opinion." Our database consisted of twenty-six MRA studies, which included 7 abnormal cases with 7 aneurysms and 19 normal cases. The vessels were first segmented from background by use of a gray-level thresholding and a region growing technique. The gradient concentration (GC) filter was then applied to the segmented vessels in order to enhance aneurysms. The initial aneurysm candidates were identified in the enhanced image by use of a gray-level thretholding. For removal of false positives (FPs), we determined four features, i.e., size, sphericity, mean value of GC image, and maximum value of GC image in each of the initial candidate regions. Finally, rule-based schemes with these features and a quadratic discriminant analysis were applied for distinction between aneurysms and FPs. The sensitivity of our method for detection of aneurysms was 100% (7/7) with 1.38 FPs per patient. Our computerized scheme may be useful in assisting radiologists for identifying correct aneurysms and for reducing the interpretation time.
社団法人電子情報通信学会の論文
2005-09-15